Traj2Video: Code for Video-Based Transportation Mode Recognition

1.4 GB of core code implements a dual-path framework for learning from trajectory-to-video representations. The repository, authored by Kunkun Fan and last updated in May 2026, is shared under a CC-BY-4.0 license on figshare. It likely contains scripts and models for converting movement data into video sequences for classification tasks.

Use Cases

• Training multiclass classifiers for transportation modes based on trajectory-to-video conversion.
• Benchmarking video-based representation learning frameworks against other methods.
• Reproducing research on dual-path neural network architectures for spatiotemporal data.

Strengths

• The 1.4 GB size suggests a substantial codebase with supporting assets.
• The CC-BY-4.0 license provides clear, permissive terms for reuse and modification.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for downstream data tasks.

Provenance

Source

Kunkun Fan

Freshness

Last updated 2026-05-16 16:20:25; freshness should be verified.